Proteoglycan Synthesis in Normal and Lowe Syndrome Fibroblasts*

Vol. 262. No. of 12, Issue

THEJOURNAL OF BIOLOGICAL CHEMISTRY

April 25,pp. 5637-5643, 1967 Printed in U.S.A.

Proteoglycan Synthesis in Normal and Lowe Syndrome Fibroblasts* (Received for publication, October 14, 1986)

Gregory S. Harper$, Vincent C. Hascall§, Masaki Yanagishitag, and William A. GahlS From the $Section on Human Biochemical Genetics, Human Genetics Branch, National Institute of Child Health and Human Development and the §Bone Research Branch, National Institute of Dental Research, National Institutes of Health, Bethesda, Maryland 20892 Lowe (oculocerebrorenal) syndrome (LS) is an X - disorder characterized by congenital cataracts, generalized renal tubular Fanconi linked disorder characterizedby congenital cataracts, hypotonia, mentalretardation,and generalized hypotonia, mental retardation, and renal syndrome (1). A range of other histological, clinical, and Fanconi syndrome. The basic defect remains unknown, biochemical findings have been reported (2-4), but no coherbut thepossibility that fibroblasts expressreduced sul- ent explanation for all aberrant findings in LS has been fation of glycosaminoglycans has been studied in sev- constructed. The observation of urinary excretion of hyposuleral laboratories. A mechanism involving overproduc- fated glycosaminoglycans (5, 6) has led to thehypothesis of a tion of an enzyme (nucleotide pyrophosphatase) active primary defect in proteoglycan sulfation. This hypothesis is againstadenosine3’-phosphate,5‘-phosphosulfate attractive since the ubiquitous involvement of proteoglycans (PAPS) has been postulated. Decreased synthesis of in basement membrane structureand function (7), tissue normally sulfated glycosaminoglycans was also re- turgor maintenance (€9, and tissue development (9) could ported. We measured the synthesis of proteoglycans explain the many manifestations of Lowe syndrome (3). and glycosaminoglycans by incorporation of r3H]gluGlycosaminoglycan synthesis andsulfation have been studcosamine and Naz36S0,into cultured fibroblasts from ied in cultured skin fibroblasts from LS patients (4, 10, 11). four LS patients and related it directly to the synthesis Fukui et al. (4) reported undersulfation of glycosaminoglycans in six normal fibroblast cultures. We found that the produced by LS fibroblasts grown under both physiologic and rate of synthesis varied greatly among the normal reduced inorganic sulfate conditions. The undersulfation was cultures (cv, 30%),but not significantly between LS confined to chondroitin and dermatan sulfate glycosaminoand the normal. The LS fibroblasts’ ability to sulfate glycans, with heparan sulfates not affected. Total glycosamiglycosaminoglycans was assayed as the amount of 3H- noglycan synthesis was lower in LS cells than in normal (4) glycosaminoglycan eluting at low ionic strength on and depended on the concentration of inorganic sulfate in the anion exchange chromatography, the amount of non- medium. Synthesis of undersulfated glycosaminoglycanswas sulfated disaccharide present in chondroitinase digests attributed to depleted PAPS pools, reportedly resulting from of labeled proteoglycans, and the ratio of ”S to 3H elevated levels of a nucleotide pyrophosphatase active against incorporation intoproteoglycans. Each parameter sug- this precursor. One study found an elevated level of this gested that the LS cells were synthesizing normally enzyme in 21 LS patients (12), with some genetic heterogesulfated glycosaminoglycans (e.g. % ADi-OS, 21 f 6 in neity. The enzyme levels in cells from heterozygotes were normal; 27 2 6 in LS). The cells’ ability to sulfate intermediate between normal and LS cells (12). glycosaminoglycans was tested under conditions of Donnelly et al. (10)found aberrant extracellular localization markedly stimulated glycosaminoglycan synthesis, by of normally sulfated glycosaminoglycans in LS, as well as treating the cultureswith a 8-D-xyloside. LS and nor- undersulfation, but highlighted the inconsistency of these mal cells responded to the treatment by elevating the observations with the proposed PAPS defect. They suggested rate of synthesis of normally sulfated glycosaminoglya membrane defect of the Golgi complex,but found no abnorcans (3.5-&fold in normal, 3-7-fold in LS). Nucleotide pyrophosphatase activities were found to be elevated mality in total glycosaminoglycan synthesis in LS cells. All the above experiments involved comparisons between in each of our four LS cell strains as in the previous cultures from a single LS patient with those from a single studies, excluding genetic heterogeneity as an explanation for our findings. We conclude that LS fibro- normal. In this paper, we assayed the effects of individual blasts do not express defects in sulfationof glycosami- variation and genetic heterogeneity upon proteoglycan synthesis by measuring proteoglycan production and sulfation in noglycans or in synthesis of proteoglycans. six normal fibroblast cultures and in four LS cultures derived from three unrelated families. The studies were performed under conditions of reduced inorganic sulfate (100 PM) and Lowe (oculocerebrorenal) syndrome (LS)’ is an X-linked under conditions which maximally stimulated chondroitin sulfatesynthesis (1 mM P-D-xyloside) and presumably * The costs of publication of this article were defrayed in part by stressed the cellular capacity to synthesize sulfated proteoglythe payment of page charges. This article must therefore be hereby cans. The results indicated substantial variation within each marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ’ The abbreviations used are: LS, Lowe (oculocerebrorenal) syndrome; PAPS, adenosine 3’-phosphate,5‘-phosphosulfate; HA, hyaluronic acid; HPLC, high performance liquid chromatography; CS, chondroitin sulfate; DS, dermatan sulfate; ADiOS, 3-O-p(~-gluc-4-

eneuronosy1)-N-acetylgalactosamine;ADi-rS and ADi-GS, derivatives of ADi-OS with a sulfate at the 4 or 6 position of the hexosamine moiety, respectively; ADi-diS, a derivative of ADi-4S bearing an additional sulfate residue at position 2 of the A4,5-glucuronic acid moiety.

5637

5638

ProteoglycaninSynthesis

Lowe Syndrome

group of normal and mutantcultures, and were not consistent newly synthesized proteoglycans. @-D-xylosidewas added to with a defect in glycosaminoglycansulfation in LS skin fibro- some cultures to study the effect of stimulated synthesis on blasts. each of these parameters. The rates of glycosaminoglycan and proteoglycan synthesis AND RESULTS* MATERIALS,METHODS, varied considerably among normal cultures (Table 111), but the range of rates encompassed those of the LS cultures. DISCUSSION Hence, there was no significant difference between LS and We studied proteoglycan synthesis in LS cells because normal. Compartmental localization of glycosaminoglycans previous clinical and histological data have suggested a defect was determined from synthetic rate andchondroitinase digesin extracellular matrix structures (3, 14, 15). A defect specif- tion data. The normal and LS cell strains secreted similar ically in glycosaminoglycan or proteoglycan synthesis was proportions of the total incorporated counts into themedium indicated by biochemical studies using skin fibroblasts (4, 10, and cell layer fractions. Both distributed chondroitin/derma11). These cells produce at least five different families of tan sulfate glycosaminoglycans into both the medium and cell proteoglycans, as well as theglycosaminoglycan HA. Heparan layer fractions and heparan sulfate primarily into the cell sulfate proteoglycans have been found primarily on the cell layer fraction. These estimates arein accord with the findings surface, with a closely related species appearing in themedium of others (28). The extentof sulfation of the glycosaminogly(16-18). Dermatan sulfate and chondroitin sulfate proteogly- cans synthesized by the normal and mutant cells was evalucansare found primarily in the medium and pericellular ated by three methods: comparison of the rateof synthesis of compartments (16). The larger of the two species contains species appearing in the HA fraction; comparison of the ratio glucuronic acid-rich chains, the smaller contains relatively of 35Sto 3H incorporation into the proteoglycan species; and iduronic acid-rich chains (19). The total proteoglycan syn- comparison of the proportion of ADi-OSdisaccharide in chonthetic ratein cultured fibroblasts varies with the tissue source droitinase digests of the glycosaminoglycans. Each parameter of the cells (20), the in vitro age of the culture(21), the exhibited a range of values for normals, but again the LS substratum used to plate the cells (22), and as yet unidentified cultures fell within this range. factors (23). Despite this complexity, the fibroblast provides /%D-xyloside treatment of fibroblasts induces a 5-8-fold a valuable system for the study of specific defects in proteo- increase in 35Sincorporation into medium fraction CS and glycan metabolism. For example, the degradation pathways DS glycosaminoglycans (29). The xyloside acts as anartificial of glycosaminoglycanshave been elucidated using fibroblasts chain initiator in the Golgi (30-33) and hence relieves the from patients with mucopolysaccharide storage disorders (24). rate limitation normally exerted by core-protein supply. The We compared several synthetic parameters for six normal syloside-initiated glycosaminoglycans are secreted into the and for four LS fibroblast strains from three unrelated fami- medium. In our synthetic rate experiments, LS cells may not lies. Synthesis of intact proteoglycans, rather than isolated have expressed defective sulfation of glycosaminoglycans beglycosaminoglycans(4), was studied since the presence of the cause sulfate precursor pools were not sufficiently depleted. core protein is essential to the localization and function of p-D-Xyloside treatment should increase the flux of sulfate the proteoglycans (25) and hence potentially involved in the through the precursor pool and supply machinery and might expression of the metabolic defect. Our preliminary experi- be expected to uncover a subtle defect in the LS cells. The ments involved analytical fractionation of proteoglycans la- rate a t which LS cells incorporated [35S]sulfateinto medium beled in large scale cultures under approximately physiologic glycosaminoglycanswas stimulated 2.8-7-foldbyxyloside sulfate conditions (1100 phi). While these experiments sug- treatment (Tables I11 and IV), yet there was no apparent gested that proteoglycan synthesis and sulfation were not change in any of the three sulfation parameters. It seems defective in LS, it was clear that an estimate of the normal likely that even if PAPS pools weredepleted in LS cells, there variability of each synthetic parameterwas required. Further, remained sufficient functional capacity to supply normal since Fukui et al. (4) suggested that the sulfation defect was amounts of sulfate for the glycosaminoglycans. poorly demonstrated under repleted sulfate conditions, subThe possibility that genetic heterogeneity within the clinisequent experimentsinvolved reduced sulfate media. We em- cally defined syndrome accounts for our cell strains not exployed 100 KM sulfate as the reduced sulfate condition rather hibiting the sulfation defect was addressed by measuring the than 30 PM (4), because Sobue et al. (26) and Humphries et activity of nucleotide pyrophosphatase. This enzyme has been al. (27) found that very low sulfate media inhibited normal found to be elevated by approximately &fold in fibroblasts sulfation of the chondroitin glycosaminoglycans. The latter from 21LS patients and, consistent with a gene-dosage effect, group also found that heparan sulfation was spared, which by approximately &fold in obligate heterozygotes (12). Nusuggests that some of the characteristics previously attributed cleotide pyrophosphatase was also elevated in our four LS cell to LS cells (which included sparing of heparan sulfation) may strains, so genetic heterogeneity with respect to this enzyme have resulted from the stringent cultureconditions employed. could not explain the absence of a glycosaminoglycan sulfaSmall scale cultures were employed to facilitate comparison tion defect in our studies. of several cell strains simultaneously. Three parameters were Previous studies have shown the enzyme to have broad extensively studied rateof glycosaminoglycan and proteogly- specificity, hydrolyzing UDP-Glc, ATP, and NAD, as well as can synthesis; localization of synthetic products into either PAPS (34). To exclude the possibility that UDP-hexosamine medium or cell layer fractions; and extent of sulfation of pools were significantly depleted by the over abundance of Portions of this paper (including “Materials,” “Methods,” “Re- the enzyme in LS cells, the specific activity of 3H in the pool sults,” Tables I-v, Figs. 1-4, and Footnotes 3 and 4) are presented in was estimated by measurement of the 35S/3Hratio in purified miniprint at the end of this paper. Miniprint is easily read with the disaccharides (Table V). The range of normal included the aid of a standardmagnifying glass. Full size photocopies are available values found for the LS cells. from the Journal of Biological Chemistry, 9650 Rockville Pike, Beto[“c] Donnelly et al. (10) reported a ratio of [35S]thesda, MD 20814. Request Document No. 86M-3560, cite the au- glucosamine incorporation into glycosaminoglycans which,at thors, and include acheckor money orderfor $10.00 per set of photocopies. Full size photocopies are also included in the microfilm 72 h of incubation, was lower than normal in theLS intracellular and pericellular fractions and higher than normal in the edition of the Journal that is available from Waverly Press.

ProteoglycaninSynthesis

Lowe Syndrome

5639

medium (Fig. 3 of Ref. 10). Fukui et al. (4), using low sulfate 13. Yamashina, I., Funakoshi, I., Fukui, S., and Yano, T. (1984) Am. J. Hum. Genet. 36, 24S, abstr. 065 medium for 72 h, noted only a slight decrease in the 35S/3H 14. Witzleben, C. L., Schoen, E. J., Tu, W. H., and McDonald, L. W. ratio in chondroitin sulfate and dermatan sulfate of LS me(1968) Am. J. Med. 4 4 , 319-324 dium compared with normal. Therefore, the claim of defective 15. Suschke, J., and Murken, J.-D. (1969) Monutsschr Kinderheilkd. 117,412 sulfation in LS rested heavily upon differences in the 35S/3H ratio measured for the cell-associated fractions, since the 16. Vogel,K. G., and Peterson, D.W. (1981) J. Bid. Chem. 2 5 6 , 13235-13242 medium fraction did not reflect these differences (Table I of 17. Carlstedt, I,, Coster, L., Malmstrom, A., and Fransson, L.-A. Ref. 4). The present studies suggest that fibroblasts express (1983) J. Biol. Chem. 2 5 8 , 11629-11635 little specificity in localization of CS/DS proteoglycans (ie. 18. Coster, L., Malmstrom, A., Carlstedt, I., andFransson, L.-A. both medium and cell layer fractions contained proteoglycans (1983) Biochem. J. 215,417-419 of similar disaccharide content), andhence synthetic defects 19. Malmstrom, A., Carlstedt, I., Aberg,L., and Fransson, L.-A. (1975) Biochem. J. 151,477-489 are unlikely to be limited to only one extracellular compart20. Conrad, G . W., Hamilton, C., and Haynes, E. (1977) J . Bwl. ment.In fact, the entire difference inchondroitinsulfate Chem. 252,6861-6870 sulfation noted by Donnelly et al., Le. decreased sulfation of 21. Vogel,K. G., Kendall, V.F., and Sapien, R. E. (1981) J. Cell. glycosaminoglycans in the pericellular fraction of LS cells Physiol. 107,271-281 (Table I1 of Ref. lo), depends upon the reproducible separa- 22. Gallagher, J. T., Gasiunas, N., and Schor, S. L. (1983) Biochem. J. 2 1 5 , 107-116 tion of pericellular from the cell layer fractions by trypsini23. Vogel, K. G., and Campbell, K. (1986) Connect. Tissue Res. 1 5 , zation. 187-198 We conclude that previously demonstrated differences in 24. Neufeld, E. F., and Cantz, M. (1973) in Lysosomes and Storage glycosaminoglycan synthesis or sulfation between a single Diseases (Hers, H. G., and Hoof, F. Van, eds) pp. 261-275, normal and a single LS fibroblast strain may reflect serenAcademic Press, Orlando, FL dipitous choices of individual cultures for study. Despite ele- 25. Hassell, J. R., Kimura, J. H., and Hascall, V. C. (1986) Annu. Rev. Biochem. 55,539-567 vated nucleotide pyrophosphatase activities, LS fibroblasts in Sobue, M., Takeuchi, J., Ito, K., Kimata, K., and Suzuki, S. 26. culture demonstrated no obvious defects in glycosaminogly(1978) J. Biol. Chem. 2 5 3 , 6190-6196 can sulfation, displayed normal HA and proteoglycan synthe- 27. Humphries, D.E., Silbert, C. K., andSilbert, J. E. (1984) in sis in low sulfate medium, were stimulated in a normal way Proceedings of the VIIZ International Symposium on Glycoconby p-D-xyloside,and synthesized normal proportions of chonjugates (Davidson, E. A., Williams, J. C., and DiFerrante, N. M., eds) p. 305, Praeger Publishers, New York droitin sulfate disaccharides. 28. Kresse, H., Von Figura, K., Buddecke, E., and Fromme, H. G . Acknowledgments-We wish to thank Drs. Zebrower, Kieras, and Brown for supplying a preprint of a paper describing the HPLC disaccharide technique, Isa Bernardini and Diana Nomanbhoy for their technical assistance, and Colleen Genovese for assistance in typing this manuscript. REFERENCES 1. Lowe, C. U., Terrey, M., and MacLachlan, E. A. (1952) Am. J. Dis. Child. 8 3 , 164-184 2. Kieras, F. J., Houck, G . E. Jr., French, J. H., and Wisniewski, K. (1984) Biochem. Med. 3 1 , 201-210 3. Wisniewski, K. E., Kieras, F. J., French, J. H., Houck, G. E. Jr., and Ramos, P. L. (1984) Ann. Neurol. 16,40-49 4. Fukui, S., Yoshida, H., Tanaka, T., Sakano, T., Usui, T., and Yamashina, I. (1981) J. Bwl. Chem. 2 5 6 , 10313-10318 5. Akasaki, M., Fukui, S., Sakano, T., Tanaka, T., Usui, T., and Yamashina, I. (1978) Clin. Chin. Acta 89, 119-125 6. Hayashi, S., Nagata, T., Kimura, A., and Tsurumi, N. (1978) Tohoku J. Exp. Med. 126,225-234 7. Kanwar, Y. S., Linker, A., and Farquhar, M. G . (1980) J. Cell Bwl. 86,688-693 8. Comper, W. D. (1984) in Edema (Staub, N. C., and Taylor, A. E., eds) pp. 229-262, Raven Press, New York 9. Toole, B. P. (1982) Connect. Tissue Res. 10,93-100 10. Donnelly, P. V., Reed, P., and DiFerrante, N. (1984) Connect. Tissue Res. 13,89-98 11. Yokoi, T., Taniguchi, N., and Ikawa, K. (1982) J. Lab. Clin. Med. 100,461-468 12. Yamashina, I., Yoshida, H., Fukui, S., and Funakoshi, I. (1983) Mol. Cell. Bwchem. 52,107-124

(1975) Hoppe-Seykr’s Z. Physiol. Chem. 356,929-941 29. Dietrich, C. P., Armelin, H.A., Nogueira, Y. L., Nader, H. B., and Michelacci, Y. M. (1982) Biochim. Biophys. Acta 717,387397 30. Okayama, M., Kimata, K., and Suzuki, S. (1973) J. Biochem. (Tokyo) 74,1069-1073 31. Robinson, H. C., Brett, M. J., Tralaggan, P. J., Lowther, D.A., and Okayama, M. (1975) Biochem. J. 148,25-34 32. Helting, T., and Rod&, L. (1969) J. Biol. Chem. 244,2790-2798 33. Galligani, L.,Hopwood, J., Schwartz, N.B., and Dorfman, A. (1975) J. Bwl. Chem. 250, 5400-5406 34. Yoshida, H., Fukui, S., Yamashina, I., Tanaka, T., Sakano, T., Usui, T., Shimotsuji, T., Yabuuchi, H., Owada, M., and Kitagawa, T. (1982) Biochem. Biophys. Res. Commun. 107, 11441150 35. Carrino, D.A., and Caplan, A. I. (1982) J. Bwl.Chem. 257, 14145-14154 36. Yanagishita, M., and Hascall, V. C. (1983) J . Biol. Chem. 258, 12857-12864 37. Mason, R. M., Kimura, J. H., and Hascall, V. C. (1982) J. Biol. Chem. 257, 2236-2245 38. Zebrower, M. E., Kieras, F. J., and Brown, W. T. (1986) Anal. Biochem. 157,93-99 39. Lundquist, P., Martensson, J., Sorbo, B., and Ohman, S. (1980) Clin. Chem. 2 6 , 1178-1181 40. Lowry, 0.H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275 41. Yanagishita, M., and Hascall, V. C. (1985) J. Biol. Chem. 2 6 0 , 5445-5455 42. Levinson, C. (1978) J. Cell. Physiol 95, 23-32 43. von Dippe, P., and Levy, D. (1982) J. Bid. Chem. 257, 43814385

Continued on next page.

5640

Proteoglycan Synthesis in Lowe Syndrome

MTERIALS

Culture designation

Patient

mtllnt fhla study

Cell Repository

Age

Culture PDL.

NI

3349

10

6

NZ

3348

IO

6

N3

5008

IO

9

5

I

n4 N5

2938A

3

6

K6

0497

4

10

3

10

11

10

RESULTS Ll

u L3

u

1676

IO

5

11

6

ProteoglycaninSynthesis

LS

-8"

-

Lowe Syndrome

5641

LI

n.5

0.9

0.5

0.4

0.4

0.04

0.07

L2

14.8

4.5

2.5

4.6

2.8

0. LO

0.11

L3

5.1

7.1

3.1

6.6

2.8

0.09

0.09

L4

1.7

3.4

2.9

3.3

2.8

0.10

0.10

so

9.3

A

4.0

4.0

A

2.6

2.3

1.2

3.7

5

2.6

2.2

t

1.2

0.08

t 0.03

0.09

0.02

ProteoglycaninSynthesis

5642 Analysis of Chondroitin Sulfare Disaccharides

O'

Lowe Syndrome Synthetic Rates in the Presence of E-D-Xyloside

a

70 I

a

6o 50

I

es

40

30

-

-

-

-

20 10

3 ,

HS

80

ADi-diS

ADi-4S

ADi-6s

3

ADi-OS

5 I

70

X

I

a

n 0

n

-

6o

73

3

50 I

R

-J

I-

40

30 20

10

20

30

40

Fraction Number

10

HS

ADi-diS

ADi-6S

LS

&an

-

ADi-4S

ADi-OS

L,

7.8

2.2

u.4

2.8

".4

0.13

0.12

LZ

16.7

11.8

1.9

16."

2.3

0.lb

0.11

1

6.6

16.7

1.3

18.3

2.1

0.11

0.10

14

2.8

10.6

2."

16.6

2.7

U.lb

so

8.4

2 10.1 5.9

6.0

1.7

0.9

13.4

t 7.2

1.9 f 1 . 0

0.13

0.02

".I&

0.12

:0.02

50

80 70

60

T T

50 40 T

30 20 10

AD-6S

ADi-4S

ADi-OS